Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
LLM Copyright/Plagiarism filters trivially bypassed with 0% detection [pdf] (paperclipmaximizer.ai)
1 point by ycombiredd 2 days ago | hide | past | favorite | 2 comments




I evaluated the Copyleaks copyright/plagiarism detection platform’s ability to detect phonetically and semantically equivalent text that differs orthographically from the original source using tools I made and link to in the paper.

Copyleaks’ own marketing materials cite accuracy rates above 99 % on “paraphrased and disguised” text. In contrast, our trials yielded detection rates as low as 0 %, with multiple transformed works passing undetected despite maintaining near-verbatim semantic and phonetic equivalence to the originals.


Here's a Medium version if you don't like clicking PDFs.

https://medium.com/@scott.vr/part-of-your-world-e6a0d78d46b9




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: